This invention relates to performance management of a computer system which includes a storage system, and more particularly, to a method of monitoring performance of a computer system by using an agent.
Storage networks in which a plurality of computers access an integrated storage system via a network are available today. These storage networks enhance the utilization efficiency of storage systems ever increasing in scale and help reduce the management cost as well, which are making storage networks widely popular as architecture for data centers. In a storage network environment, there is a possibility that the processing loads of application systems interfere with one another in a component (e.g., a network system or a storage system) shared by a plurality of computers. The load interference could cause a performance problem (lowering of the performance of the application systems, for example) in the application systems. A computer system to which a storage network as those described above is introduced therefore needs to collect performance information on resources constituting the network in a comprehensive manner in order to monitor and adjust the performance of the application systems.
An example of a performance monitoring technique in a networked computer system is disclosed in U.S. Pat. No. 6,505,248. According to U.S. Pat. No. 6,505,248, performance is managed by an agent arranged for each of resources which are to be monitored, and management software for uniformly managing performance information of the entire system.
Each of the agents obtains the performance information by directly communicating with the monitoring target. Specifically, the agent executes polling with respect to the monitoring target to obtain a metrics value of resources.
On the other hand, the management software collects and stores the pieces of performance information obtained by the agents, and provides the stored pieces of performance information according to a request from an administrator or the like. Accordingly, it is possible to uniformly manage operation states and performance of the resource constituting the computer system.
When the size of the uniformly managed computer system increases, the number of resources to be monitored also increases. Along with an increase in the number of resources, an amount of the performance information which the management software must obtain also increases, whereby it is necessary to process a large amount of performance information. JP 2005-157933 A discloses a method of automatically adjusting a target range from which performance information is to be obtained and an obtaining frequency based on the information obtained by the management software.